Timeliness is one of the key indicators of cancer surveillance data quality, as delayed reporting of cases results in an underestimation of the cancer rate in a population. The purpose of this paper is to assess temporal trends in reporting delay by cancer site from 1999-2010. Using data from the Surveillance, Epidemiology, and End Results (SEER) 9 cancer registries and the New York State Cancer Registry, I calculated short-, medium-, and long-term delay for the most common cancer sites for each year and identified the linear trend. Nearly all sites showed a decrease in delay over the period, and many showed a statistically significant decrease. The decrease in delay was more pronounced in the New York State data. These findings reflect long-term improvement in the timeliness of cancer reporting, but there remains room for improvement. Leukemia and myeloma are especially problematic, as these sites are heavily dependent on reporting by private physicians.